Multiplexed discovery of sequence polymorphisms using base-specific cleavage and MALDI-TOF MS

Abstract

The completion of the Human Genome Project provides researchers with a reference sequence that covers about 99% of the gene-containing regions and is more than 99.9% accurate. Sequence drafts and completed sequences for several other species are also available to researchers worldwide. The ongoing effort to provide more and more genomic reference information now enables the detection of deviations from this 'genetic blueprint'. Comparative sequencing projects will play a major role in elucidating the meaning of the genetic code and in establishing a correlation between genotype and phenotype. As part of this effort, a number of projects will focus on distinct functional aspects, like resequencing of exons or HLA determining regions. Typically these target regions are short in length and their analysis does not require long read length. To find an efficient solution for these applications, we developed a novel method that allows simultaneous analysis of multiple independent target regions (Multiplexed Comparative Sequence Analysis) by employing base-specific cleavage biochemistry and MALDI TOF-MS analysis.

Figure 1

(a) Schematic representation of multiplexed sequence polymorphism discovery enabled through base-specific cleavage and MALDI-TOF MS. Single-stranded RNA molecules are generated from the PCR product and are then cleaved to completion base-specifically. Cleavage products are analyzed in parallel with MALDI-TOF MS. The cleavage products are independent of each other and hence can be generated from one or multiple templates. (b) Mass spectra of the base-specific cleavage reaction for the three uniplex reactions tested compared with the mass spectrum derived from triplexing the same amplicons. All mass signals present in the uniplex reaction are also present in the multiplexed reaction with a homogeneous distribution of signal intensities. (c) Verification of the discovery of an A/G polymorphism in multiplexed base-specific cleavage. The lower set of three mass spectra corresponds to the three possible genotypes for the polymorphism. The upper panel shows the mass spectra of the triplexed reaction for the same polymorphism. The mass signals leading to the discovery of the A/G polymorphism are clearly identified in the multiplexed reaction.

Figure 2

Simulation results for predicted SNPs (‘known’ SNPs from ensembl database) and all theoretically possible SNPs using the reference sequence for all exons of Ensembl (build 34b). The y-axis provides the percentage of sequence change events that can be detected with base-specific cleavage (using four cleavage reactions). The x-axis provides the simulation results for overall amplicon length 200 up to 1.5 kb. Solid lines represent uniplexing results, dashed lines represent triplexed re-sequencing, dotted lines pentaplexed re-sequencing.

Figure 3

Comparison of spectra quality based on (a) confidence scores in uniplex and triplex reactions and (b) signal intensities for each individual amplicon derived from the triplexed reactions. The distribution of confidence scores and signal intensities are presented as box and whisker plots. The box extends from the first to the third quartile (interquartile range). The line inside the box marks the median and the lines extending out on either side indicate the minimum and maximum values.

Figure 4

Representative mass spectrum of a heptaplexed base-specific cleavage reaction of discovery of sequence polymorphism in the LAMR1 gene. The overall nucleotide count in this multiplex totaled to 1850 bases. All expected mass signals are marked with a dashed line. The mass signal labeled ‘*’ represents an additional signal derived from primer dimer formation, the signal labeled ‘‡’ represents an abortive cycling product and the signal labeled ‘#’ represents a sodium adduct of the cleavage product GGGAGAAGGCT at 3913 Da (+22 Da).

Figure 5

Overlay of three mass spectra showing the ΔF508-specific mass signal changes generated in the T-specific cleavage reaction of the reverse strands. The mass signal pattern is derived from a triplexed reaction encompassing exon 10, 21 and 24 of the CFTR gene. Three different individuals were analyzed, the upper mass spectrum representing a homozygous wild-type for ΔF508, a carrier of the ΔF508 mutation (middle) and an individual homoyzgous mutant ΔF508 (lower mass spectrum). Mass signals specific for the mutation are indicated with a line.